Assessing the Hydrochemistry, Groundwater Drinking Quality, and Possible Hazard to Human Health in Shizuishan Area, Northwest China

Abstract

Groundwater is an important source of drinking water, particularly in arid regions. In this study, a total of 66 groundwater samples were collected from the phreatic aquifer in the Shizuishan area, a traditional irrigation region of Ningxia. The results showed that the TDS values were above the drinking water standards for nearly 50% of the groundwater samples. The ions followed the order of Na⁺ > Ca²⁺ > Mg²⁺ > K⁺ and SO₄²⁻ > Cl⁻ > HCO₃⁻ in the groundwater. There were four dominant factors in controlling groundwater chemistry based on principal component analysis: the salinity factor, alkalinity factor, carbonate factor, and pollution factor. The high concentration of NH₄-N in groundwater was attributed to agricultural activities, but the high NO₃-N levels were mainly due to sewage or wastewater. F and As were derived from geogenic sources. Based on the result of the WQI assessment, about 40% of the samples in the central part of the study region showed unacceptable water quality for drinking, which was mainly associated with high NH₄-N, TDS, and As concentrations. The total non-carcinogenic risks of drinking the groundwater were 0.05–10.62 for adults and 0.09–20.65 for children, respectively. The order of pollutants in the groundwater in terms of their hazard to residents was: As > F⁻ > NO₃-N > NH₄-N. The carcinogenic risk values of As through oral ingestion for children and adults were 0–7.37 × 10⁻⁴ and 0–1.89 × 10⁻⁴, respectively. Chronic exposure by oral ingestion presented as the main source of susceptibility to exposure to groundwater contaminants for children.

1. Introduction

Groundwater plays an important role in supporting agricultural, domestic, and industrial water use because of its widespread distribution and relatively stable quality [1].

In many semi-arid and arid regions, groundwater is a major source of drinking water [2,3]. The impacts of human and geogenic activities on groundwater quality are of increasing concern. Geogenic sources, rapid urbanization, excessive withdrawals, improper disposal of waste, the overuse of fertilizer, etc., have caused major changes in the physical properties and chemical composition of groundwater to a great extent [4,5,6]. However, prolonged exposure to contaminants can have adverse effects on human health. Groundwater remediation is also a long and slow process.

Understanding the hydrochemistry of groundwater is vital for maintaining groundwater quality [7,8,9]. The chemistry of groundwater has evolved through the effects of precipitation, evaporation, weathering, sorption, and exchange reactions [10,11,12]. Many scholars have found by various methods that the chemical composition of groundwater is closely related to the geological environment and hydrogeological conditions [13]. For instance, Schot et al. [14] found that the hydrochemical composition of groundwater in the Gooi and Vechtstreek areas in the Netherlands was influenced by human activities, with urbanization and agricultural activities causing increased concentrations of nitrate, sulfate, and K⁺ in the groundwater. Adams et al. [15] indicated that the chemical composition of the groundwater in the study area is dominated by salinization, mineral dissolution and precipitation, cation exchange, and human activities. Zhang et al. [16] showed that the high concentrations of Fe and Mn in the Songnen Plain were associated with the unique geomorphology and reducing environment. Güler et al. [17] showed that the main factors causing water chemistry changes in the Tarsus coastal plain were water-rock interaction and nitrate pollution, seawater intrusion and salinization, geological factors, and anthropogenic zinc pollution. In this context, the exposure through drinking water to high levels of nitrogen, fluoride, and arsenic is of increasing concern to humans. Nitrogen is one of the essential elements for plant growth [18,19]. Although nitrogen fertilizer provides certain nutrients for plant growth and improves crop yields, the use of large amounts of nitrogen fertilizer accelerates the leaching of nitrogen into groundwater, which will pose a serious threat to the environment and human health [20,21].

Groundwater contaminated with high levels of arsenic and fluoride is a serious threat to the safety of drinking water. It has become a global problem [22,23,24]. The chronic ingestion of high-fluoride (>1.0–1.5 mg/L) groundwater has been shown to contribute to fluorosis and various diseases, such as cardiovascular disease, osteosclerosis, endocrine disorders, and multi-organ lesions [10]. Chronic exposure to arsenic through contaminated food and drinking water can lead to arsenic poisoning, with symptoms including peripheral neuropathy, skin lesions, diabetes, cancer, and cardiovascular disease [25,26,27]. The Water Quality Index (WQI) is a practical approach to assessing the quality of groundwater by distilling a large amount of water quality data to an index representative of regional groundwater quality [28,29,30,31]. The United States Environmental Protection Agency’s (USEPA) Human Health Risk Assessment (HHRA) (USEPA, 2004) is widely used to assess the risks of groundwater quality to human health. Integrating the HHRA and WQI can facilitate improved monitoring and groundwater quality conservation for sustainable groundwater management [32,33,34,35].

In northwestern China, water obtained from alluvial aquifers is used for drinking purposes, especially in most remote areas [36,37]. The Shizuishan area, located in northern Ningxia, is a typical arid and semi-arid region. As a traditionally agricultural region, a large amount of irrigation water infiltrated into the aquifer becomes the dominant groundwater inflow. Private hand-pumped wells are widely used in the rural part to support drinking due to their low cost and high efficiency. However, the phreatic aquifer is sensitive to human activities because of the shallow water table coupled with the high population densities. It is still a potential risk to human health to use groundwater as a source of drinking water in the study region. Therefore, the present study aims to interpret the complex groundwater hydrochemistry of the shallow aquifer in the Shizuishan area. In the present study, Principal Component Analysis (PCA) was applied to identify natural and anthropogenic factors with significant influence on groundwater hydrochemistry. The quality of groundwater and associated health risks were evaluated to provide more information for groundwater management.

2. Study Area

The Shizuishan area covers about 2241 km² and is located in the northern Yinchuan Plain, Ningxia (38°39′17″–39°23′16″ N, 106°8′14″–106°52′11″ E. Figure 1). The Yinchuan Plain has a traditional irrigation history dating back to more than 2000 years. According to the Ningxia Statistical Yearbook (2022) [38], Shizuishan has a population of about 800,000, and the main crops are rice, corn, and wheat. In the arid environment, the mean annual precipitation and potential evaporation are 179 mm and 1800 mm, respectively. Precipitation is mainly concentrated in May and October. The area has sufficient sunshine, a large temperature difference between day and night, and the annual average climate is about 10.6 °C [38,39]. Irrigation water in the area is mainly supplied by water transfer from the Yellow River, and there are many water diversion channels in the plain to transfer water through to ensure the needs of crops are met. Groundwater depth is impacted by irrigation and ranges from 1 to 3 m. Soil salinization has many major adverse effects on crop yield due to intense evaporation.

Groundwater serves as crucial support for drinking, irrigation, and industrial purposes. In general, the groundwater flow in the study area is from the southwest to the northeast. The aquifer in Shizuishan is mainly composed of Quaternary fine alluvial lacustrine deposits. From west to east, the landforms of the study area are leaning towards pluvial plain, pluvial alluvial plain, and alluvial lacustrine plain [40]. The lithology of aquifers in the pluvial plain becomes thinner from west to east. The lithology of the alluvial lacustrine plain is mainly medium fine sand [39]. From top to bottom, the aquifer system is composed of phreatic, upper confined, and lower confined aquifers with thicknesses of 20–30 m, 50–60 m, and 60–80 m, respectively. The aquitards consist mainly of clay between the adjacent aquifers. In this study, we focused primarily on the phreatic aquifer. The aquifer is recharged primarily by irrigation channels, irrigation infiltration, precipitation, and lateral inflow. Groundwater withdrawals include evaporation, discharge to drains, and artificial extraction. Many residents rely on groundwater for their drinking water.

3. Materials and Methods

3.1. Collection and Analyses of Samples

The present study collected 66 groundwater samples from private wells in April 2021 (Figure 1). The collection and processing of groundwater samples in the present study followed technical specifications for environmental monitoring. Sampling locations were recorded by using a handled GPS device. Before collecting groundwater samples, the well was flushed for 5 to 10 min to remove standing water. Water samples were stored in previously sterilized polyethylene bottles, and pH, total dissolved solids (TDS), and temperature were measured in the field using portable meters (DDBJ-350F, INESA Scientific Instrument Co., Ltd., Shanghai, China). The samples were stored at 4 °C before analysis for physicochemical parameters.

The groundwater samples were analyzed according to the methods recommended by the Ministry of Health of the People’s Republic of China (PRC) and the Standardization Administration of the PRC [41]. The laboratory of the Ningxia Geological Survey Institute analyzed all the samples. The samples were used for quality assurance/quality control (QA/QC) analysis. The sample analysis followed the Technical Guidelines for Groundwater Environmental Monitoring developed by the State Environmental Protection Administration (2004) [42].To ensure the accuracy and reliability of the results, the groundwater sample analysis was repeated three times. Flame atomic absorption spectrophotometry was used to measure the contents of K⁺ and Na⁺, while titration was used to measure those of Cl⁻, Ca²⁺, SO₄²⁻, Mg²⁺, and HCO₃⁻. Ion chromatography was used to measure the content of F^-. Spectrophotometry was used to measure the contents of NH₄–N and NO₃–N. Arsenic content was determined through hydride generation atomic fluorescence spectrometry (HG-AFS).

In this study, SPSS 20.0 was used for the principal component analysis of the groundwater samples to analyze the main controlling factors in groundwater. Piper (1944) and Gibbs (1970) diagrams and ion ratios were drawn from to study the hydrochemical types and main control factors of the groundwater by using MATLAB (version 2016b) and the Origin (version 2020) software. An IDW-to-raster interpolation was applied to spatially map the groundwater quality and health risks from oral intake based on ArcGIS (version 10.7).

3.2. Groundwater Quality Assessment

The WQI provides a comprehensive assessment of drinking water quality. Weights (w_i; between 2 and 5) have been assigned to the different individual parameters making up the WQI according to their relative concentrations and importance for drinking water quality [43,44]. The WQI is calculated through:

$$W_{\mathrm{i}}=\frac{w_i}{{\displaystyle \sum _i^n{w}_i}}$$

(1)

$$Q_i=\frac{C_i-C_{ip}}{S_i-C_{ip}}\times 100$$

(2)

$$S{I}_i=W_i\times Q_i$$

(3)

$$WQI={\displaystyle \sum _{i=1}^{n}S{I}_i}$$

(4)

where W_i represents the weight of each parameter, w_i represents the weight assigned to each parameter, n represents the number of parameters, C_i denotes the concentration of a single parameter, C_ip represents the ideal distilled parameter value (a pH of 7 and zero concentration for the remaining parameters), Q_i represents the quality rating, S_i represents the WHO standard for a single parameter, and SI_i represents the sub-index of the “ith” parameter. The quality of groundwater in the study area could be classified into five classes according to the WQI: (1) excellent; (2) good; (3) medium; (4) poor; and (5) extremely poor.

3.3. Assessment of Hazards to Human Health

The present study applied the HHRA to estimate the adverse impacts of the ingestion and assimilation of toxicants on adults and children. The non-carcinogenic hazard posed by contaminants was determined by [45]:

$$D_i=\frac{C_i\times IR\times EF\times ED}{BW\times AT}$$

(5)

$$H{Q}_i=D_i/Rf{D}_i$$

(6)

where D_i represents exposure dose by ingestion for the ith contaminant (mg/kg per day); C_i denotes contaminant concentration (mg/L); IR is the rate of ingestion (L/day); EF represents the frequency of exposure (days/year); ED represents the duration of exposure (years); BW means body mass (kg); AT is mean exposure duration (days); HQ_i represents the non-carcinogenic impact of an individual parameter; and RfD represents the non-carcinogenic reference dosage of a parameter (mg kg⁻¹ day⁻¹). The RfD for NO₃-N, NH₄-N, F⁻, and AS were 1.6, 0.9, 0.06, and 0.0003 mg/(kg·day), respectively [37,46,47].

The total hazard index (HI) represents the integrated hazard posed by contaminants in water used for drinking. A HI of > 1.0 and a HI of < 1.0 represent potential adverse human health impacts and acceptable levels of non-carcinogenic risk, respectively.

$$HI={\displaystyle \sum HQ}$$

(7)

In this study, the potential health risks of ingesting NO₃⁻, NH₄⁺, F⁻, and As in groundwater were considered. Because all residents in the study area rely on groundwater as a source of drinking water, EF was assigned a value of 365 days for both children and adults. ED was assigned to children and adults at ages 6 and 30. AT is equal to 365 days. Statistical data [48] indicated the BW of adults and children in Ningxia to be 62.5 kg and 15 kg, respectively. All parameters are listed in

Table 1. Parameters of daily dose calculation models.

Parameters	Unit	Item	Children	Adults
IR	L/d	Oral	0.7	1.50
BW	kg	Oral	15	62.5
EF	d/a	Oral	365	365
ED	a	Oral	6	30
AT	d	Carcinogenic	74.68 × 365	74.68 × 365
		Non-carcinogenic	6 × 365	30 × 365

.

The carcinogenic risk of arsenic (R_As) can be estimated as:

$$R_{As}=q_{As}{D}_{As}$$

(8)

where q_As represents the carcinogenic coefficient of arsenic ingested through drinking water (1.5 kg day mg⁻¹). The USEPA usually applies a range of target reference risks of 10⁻⁴–10⁻⁶ in the assimilation of carcinogenic toxins in drinking water, with 10⁻⁶ being generally recognized as an appropriate standard for drinking water [49]. R_As of > 10⁻⁴ represents the possibility of adverse impacts on human health posed by arsenic in groundwater.

4. Results and Discussion

4.1. Groundwater Chemistry

Table 2. Entropy weights for the parameters.

Parameters	Weight (Wi)	Relative Weight Wi
pH	3	0.063
TDS	5	0.104
K+	2	0.042
Na+	2	0.042
Ca2+	3	0.063
Mg2+	3	0.063
SO42−	3	0.063
Cl−	4	0.083
HCO3−	3	0.063
NO3–N	5	0.104
NH4–N	5	0.104
F−	5	0.104
As	5	0.104

provides a statistical summary of groundwater quality according to the analysis of groundwater samples (Table 2). The ranges of WHO and the Chinese standards for different chemicals [50,51] are considered in the assessment of the groundwater for use as drinking water.

The mean pH of the water samples was 7.59, indicating it was slightly alkaline. The TDS values ranged from 232 to 18,448 mg/L (mean of 1990 mg/L). The TDS value was above the drinking water standard (1000 mg/L) for nearly 50% of the groundwater samples [51]. The highest TDS concentrations in groundwater were found in Chengguan and Gaozhuang villages. The elevated TDS concentrations could be attributed to ion exchange, solubilization, and the extended groundwater residence time in the aquifer [52,53].

The cations in the groundwater could be ranked by mean content in the order Na⁺ > Ca²⁺ > Mg²⁺ > K⁺, with concentrations of 427, 135, 92, and 8.13 mg/L, respectively (Figure 2). The high concentration of sodium in groundwater may be associated with the dissolution of halite, Na-feldspar, and ion exchange [54]. Na⁺ was the dominant groundwater cation with a concentration ranging from 30 to 5540 mg/L, and one-third of the samples had Na⁺ concentrations higher than the WHO [51] standard of 200 mg/L. Groundwater Ca²⁺ and Mg²⁺ concentrations ranged from 24.02 to 1027 mg/L and 6.07 to 584 mg/L, respectively. The concentrations of Ca²⁺ and Mg²⁺ exceeded the WHO [51] limits (75 mg/L and 30 mg/L in three-quarters of the groundwater samples. Groundwater K⁺ concentration was relatively low, with 16% of the samples exceeding the WHO [51] limit of 10 mg/L. The major ions in the groundwater followed the trend SO₄²⁻ > Cl⁻ > HCO₃⁻ with mean concentrations of 663 mg/L, 487 mg/L, and 330 mg/L, respectively (Figure 2). The high concentrations may be due to the dissolution of sulfate minerals (such as mirabilite, gypsum, etc.) and human activities (such as domestic sewage discharge, industrial pollution, etc.) [55].

About 55% of the samples contained SO₄²⁻, which exceeded the WHO [51] and Chinese permissible limit [50] (250 mg/L). The concentrations of Cl⁻ and HCO3⁻ in 26% and 43% of the groundwater samples, respectively, exceeded the respective WHO [51] and Chinese permissible limits [50] (250 g/L and 300 mg/L). Most of the study region had high HCO₃⁻ concentrations in the groundwater, and only some villages (including Yuanyi, western parts of Chonggang, Xinghai, and Yanzidun; Figure 3i) along the western border had low HCO₃⁻ concentrations.

In the traditional irrigated area, the nitrogen concentration in groundwater can better indicate the anthropogenic effects [56,57,58]. The concentrations of NO₃-N in the groundwater ranged from BDL (below the detection limit) to 62.15 mg/L (Figure 2). About 20% and 5% of the samples had NO₃-N concentrations higher than the WHO [51] and Chinese [50] drinking water standards of 10 mg/L and 20 mg/L, respectively. NH₄-N concentrations ranged between 0–4.7 mg/L (Figure 2). Nearly 23% of the samples had high NH₄-N levels, exceeding the threshold of the WHO [51] and Chinese standard [50] (0.5 mg/L). Nevertheless, NH₄-N was the dominant nitrogen pollutant in the study region due to its wide spatial distribution.

Fluoride is a dominant trace element in groundwater [59,60]. High concentrations of groundwater fluoride can result in fluorosis of the teeth and skeleton, including teeth discoloration and ligament deformation [61,62,63]. Groundwater fluoride concentrations ranged from 0.06 to 1.56 mg/L (mean of 0.58 mg/L). Among the 14 groundwater samples, 2 samples in the Lihe, Chonggang, and Xinghai villages (Figure 3l) contained higher fluoride levels than recommended by the WHO [51] and Chinese standards [50] of 1.5 and 1.0 mg/L, respectively. The high fluoride content may be due to the dissolution of fluorine-containing minerals, such as fluorite [48].

According to the WHO [51] and the Chinese standard [50], a concentration of As above 0.01 mg/L may hurt human health [62,63,64]. In the study region, the concentration of As was in the range of 0 to 0.131 mg/L. While the water quality parameters of most of the groundwater samples were within the drinking water standards, 10 samples from the central region contained levels of As that were higher than the permissible limit.

4.2. Hydrogeochemical Facies

The present study used a Piper [65] trilinear diagram (Figure 4) to illustrate the ion composition and chemical evolution of the groundwater. Groundwater cations were concentrated in the center and right corners of the cation triangle diagram. Groundwater cations indicated the groundwater types, namely a non-dominant type and a Na-dominant type. Groundwater anions showed a relatively dispersed distribution, with partial concentration in the SO₄²⁻ and HCO₃⁻ terminals. The facies of groundwater hydrochemistry in the study area were predominantly SO₄·Cl-Ca·Mg, and SO₄·Cl-Na. Some samples distributed along the Helan Mountains belonged to the HCO₃⁻ Ca·Mg type.

4.3. Processes Regulating Groundwater Hydrochemistry

Principal Component Analysis (PCA) is a linear dimensionality reduction technique that reduces the number of variables and retains much more information [53]. The present study performed a PCA to identify the dominant factors regulating groundwater hydrochemistry in the study area. For the groundwater samples in this study, PC1 (43.40%), PC2 (13.63%), PC3 (13.18%), and PC4 (11.13%) could explain 81.34% of these 13 variables, which means that the principal component analysis was reliable (Figure 5).

PC1 had a high contribution of TDS, K⁺, Na⁺, Mg²⁺, Cl⁻, and SO₄²⁻ (r = 0.97, 0.92, 0.96, 0.82, 0.94, and 0.96, respectively). These ions were the major constituents of the TDS. According to the correlation analysis, there were significant (p < 0.05) positive correlations between the TDS and K⁺, Na⁺, Mg²⁺, Cl⁻, and SO₄²⁻ (r² = 0.88, 0.97, 0.90, 0.98, and 0.98, respectively Figure 6). This indicates that the source of these ions may be derived from the dissolution of gypsum, halite, and dolomite [23,66]. The strong positive correlation demonstrated the dominant role of water-rock interaction in controlling groundwater hydrochemistry. Therefore, this component is referred to as the salinity factor.

In PC2, there were strong positive loadings values of groundwater with pH and As (r = 0.83 and 0.55, respectively), and negative loadings of NO₃-N (r = −0.72). The level of the pH is an important factor for As accumulation in groundwater. Groundwater is weakly alkaline, which is helpful for As adsorption [67]. It has been shown that elevated groundwater pH affects the dissolution of As minerals due to chemical interactions between the underlying aquifer layer and the overlying water. This is due to the negative charge on the surface of sediment particles with increased pH, which forms electrostatic repulsion with arsenic-containing anions. As a result, arsenic in the adsorbed state can be desorbed and released into the groundwater [68]. Combined with the scattered distribution of high-NO₃-N groundwater and the negative relationship between pH and NO₃-N (r² = −0.42, p < 0.05, Figure 6), it is reasonable to assume that the possible source of NO₃-N may be derived from domestic and industrial wastewater infiltration. Therefore, this component is referred to as the alkalinity factor.

PC3 showed significant loadings for Ca²⁺, Mg²⁺, and HCO₃⁻ (r = 0.61, 0.50, and 0.75, respectively). These components were attributed to carbonate dissolution/precipitation. Positive significant (p < 0.05) correlations were found between Ca²⁺ and HCO₃⁻ and between Mg²⁺ and HCO₃⁻ (r² = 0.40 and 0.56, respectively, Figure 6), indicating that calcite and dolomite may be the major sources of those ions (Reaction 1 and Reaction 2). Therefore, this component is termed a carbonate factor.

$${\mathrm{CaCO}}_3{+\mathrm{CO}}_2{+\mathrm{H}}_2\mathrm{O}\to {2\mathrm{HCO}}_3{}^{-}{+\mathrm{Ca}}^{2+}$$

(R1)

$${\mathrm{CaMg}(\mathrm{CO}}_3{)}_2{+2\mathrm{CO}}_2{+2\mathrm{H}}_2\mathrm{O}\to {4\mathrm{HCO}}_3{}^{-}{+\mathrm{Ca}}^{2+}{+\mathrm{Mg}}^{2+}$$

(R2)

The PC4 had high positive loadings of NH₄-N (r = 0.71) and As (r = 0.51), and had negative loadings of F (r = −0.70). According to research [68,69,70], the occurrence of NH₄-N and As in groundwater is mostly detected in anaerobic environments in the northern part of the Yinchuan Plain. Positive relationships between NH₄-N and As (r² = 0.3, p < 0.05, Figure 6) in groundwater can demonstrate the controlling factor of redox conditions in hydrochemistry. Wu et al. [39] found that fluorite tends to dissolve in groundwater. Fluorite is the main source of fluoride in groundwater. PC4 contained the predominant groundwater contaminants in the study region. Therefore, this component is considered a pollution factor.

4.3.1. Water-Rock Interaction

Gibbs [71] proposed a model to better understand the mechanisms that regulate groundwater chemistry. This diagram reflects the influences of water-rock interactions, evaporation, and precipitation on groundwater hydrochemistry. The Gibbs curve shows that evaporation and rock weathering are the main driving forces, which are related to the complex geochemical mechanism of the study area. Other studies have made similar analyses [72]. The Gibbs diagram has been widely used in hydrogeochemical research [54,73]. Due to the hydrogeochemical facies shown in Figure 7, the majority of samples were positioned in the evaporation- and rock-dominant regions. The shallow depth of the groundwater facilitated the infiltration of irrigation water into the groundwater, resulting in evaporation, which is also the primary factor influencing groundwater chemistry. Since the study area is arid/semi-arid, groundwater sampling points in the present study were positioned along long distances from areas in which atmospheric precipitation regulates groundwater quality.

The Na⁺: Cl⁻ (meq/L) ratio is widely used to identify the main sources of Na⁺ [74]. Na⁺: Cl⁻ = 1 when groundwater Na⁺ is derived from halite dissolution. However, if the ratio is greater than 1.0, the Na⁺ may have another source, such as the dissolution of silicate rocks or cation exchange [75]. As shown in Figure 8a, most of the sampling points were near or to the left of the 1:1 ratio line. Therefore, the groundwater Na⁺ was mainly influenced by the dissolution of halite and also influenced by other sources (Reaction 3 and Reaction 4).

$$\mathrm{NaCl}\to {\mathrm{Na}}^{+}{+\mathrm{Cl}}^{-}$$

(R3)

$${2\mathrm{NaAlSi}}_3{\mathrm{O}}_8{+2\mathrm{CO}}_2{+11\mathrm{H}}_2\mathrm{O}\to {\mathrm{Al}}_2{\mathrm{Si}}_2{\mathrm{O}}_5{\left(\mathrm{OH}\right)}_4{+2\mathrm{Na}+4\mathrm{H}}_4{\mathrm{SiO}}_4{+2\mathrm{HCO}}_3^{-}$$

(R4)

Ca²⁺/SO₄²⁻ is expected to be 1:1 when Ca²⁺ and SO₄²⁻ are derived from gypsum dissolution [76,77]. As shown in Figure 8b, most of the sampling points were positioned above the 1:1 line, indicating excess SO₄²⁻ and/or insufficient Ca²⁺ in groundwater. Cation exchange may reduce the Ca²⁺ concentration in the study region. Under those conditions, a moderate positive correlation was observed between concentrations of Ca²⁺ and SO₄²⁻ (r² = 0.51, p < 0.05).

In addition, the ratio (Ca²⁺ + Mg²⁺): (HCO₃⁻ + SO₄²⁻) can be helpful for determining groundwater sources of Ca²⁺ and Mg²⁺. If the ratio is approximately 1.0, carbonate and sulfate rock dissolution is the main source of groundwater Ca²⁺ and Mg²⁺ [78]. In Figure 8c, most of the samples fell below the line. This demonstrates that the concentrations of HCO₃⁻ and SO₄²⁻ have obvious advantages for concentrations of Ca²⁺ + Mg²⁺. The groundwater content of these ions may be regulated by the dissolution of both carbonate and gypsum. In addition, the cation exchange process involving the release of Na⁺ and the uptake of Ca²⁺ or Mg²⁺ can cause a deficit in groundwater Ca²⁺ +  Mg²⁺.

Schoeller [79] proposed the chlor-alkali index (CAI-I and CAI-II) to measure the ion exchange reaction between groundwater and aquifers. If the chlor-alkali index is negative, it means that the amounts of Ca²⁺ and Mg²⁺ in the groundwater is decreasing, while those of Na+ and K⁺ are increasing. Conversely, Ca²⁺ and Mg²⁺ increased, and Na⁺ and K⁺ decreased [53,72]. The above observations of the cation exchange process were further confirmed by the results of the chlor-alkali indices. As shown in Figure 8d, groundwater CAI-I and CAI-II were negative in most of the samples, ranging from − 4.76 to 0.61 and −0.65 to 0.70, respectively. This result indicated the occurrence of positive cation exchange reactions in all the sampling areas. In the study region, groundwater Ca²⁺ displaces Na⁺ from the host rock during positive cation exchange, increasing concentrations of groundwater Na⁺ (Reaction 5).

$$CAI-\mathrm{I}=\left[C{l}^{-}-(N{a}^{+}+K^{+})\right]/C{l}^{-}$$

(9)

$$CAI-\mathrm{II}=\left[C{l}^{-}-(N{a}^{+}+K^{+})\right]/(S{O}_4^{2-}+HC{O}_3^{-}+C{O}_3^{2-}+N{O}_3^{-})$$

(10)

$${2\mathrm{NaX}+\mathrm{Ca}}^{2+}\to {\mathrm{CaX}}_2{+2\mathrm{Na}}^{+}$$

(R5)

4.3.2. Anthropogenic Activities

NO₃ and NH₄ are the main forms of nitrogen in groundwater. However, the spatial distributions of NO₃-N and NH₄-N (Figure 3j,k) indicate point and non-point contamination. The positive relationship between NH₄-N and K⁺(r² = 0.42, p < 0.05) in Figure 6 can better explain the effect of fertilizer application. As a traditional irrigated region, nitrogen pollution in groundwater is caused by the overuse of fertilizer and flood irrigation. The average fertilizer application amount per hectare of arable land in Ningxia reached 821.81 kg, and the average fertilizer application rate was much higher than the national average (339 kg/ha) [38]. Rahman et al. [56] verified the positive correlation between NH4-N and K+ in groundwater due to intensive agricultural activities. However, when the groundwater is in a reducing environment, nitrate is reduced to nitrite nitrogen and then to ammonia nitrogen under the action of parthenogenic anaerobic bacteria [69]. The anaerobic conditions determined the predominance of NH₄-N in the groundwater. This observation is supported by the results of Chen et al. [40] in which high groundwater Fe and As concentrations were observed, which was further verified by the redox condition.

4.3.3. Geogenic Sources

Minerals containing arsenic and fluoride, such as fluorite, apatite, biotite, muscovite, hornblende, etc., are widely considered to be the sources of groundwater arsenic and fluoride [80]. The Shizuishan area contains a large amount of minerals including silica, coal, and pyrite. Scorodite (FeAsO₄·2H₂O) is the main component of pyrite, and thus, it is considered a potential source of arsenic in groundwater [81]. Based on the chemical equilibrium in Reaction 6, scorodite may be a possible source of arsenic in groundwater [82]. The dissolution of CaF₂ (fluorite) is believed to be the main process leading to groundwater fluoride levels in Shizuishan (Reaction 7). This is consistent with previous research in Yinchuan Plain [36,39,76]. The high concentration of HCO₃⁻ in groundwater can control the solubility of fluoride, thereby releasing the concentration of fluoride into groundwater [83,84]. The arsenic and pH in the study region displayed an excellent positive correlation (r² = 0.42, p < 0.05). It made the alkaline environment the major controlling mechanism for the leaching of arsenic into groundwater. In the arid region, evaporation is beneficial to enhancing the concentration of arsenic in groundwater.

$${\mathrm{FeAsO}}_4\bullet {2\mathrm{H}}_2\mathrm{O}\to {\mathrm{Fe}}^{3+}{+\mathrm{AsO}}_4^{3-}{+2\mathrm{H}}_2\mathrm{O}$$

(R6)

$${\mathrm{CaF}}_2{+2\mathrm{HCO}}_3^{-}\to {\mathrm{CaCO}}_3{+2\mathrm{F}}^{-}{+\mathrm{H}}_2{\mathrm{O}+\mathrm{CO}}_2$$

(R7)

4.4. Groundwater Quality Assessment

Based on the weight given for each parameter, the relative weights were calculated (Table 2). The relative weights of TDS, Cl⁻, NO₃-N, NH₄-N, F⁻, and As were greater than other parameters, indicating that these were parameters with dominant effects on groundwater quality.

WQI is used to assess groundwater quality [43,44]. A WQI of < 50, 50 < WQI < 100, 100 < WQI < 200, 200 < WQI < 300, and > 300 indicate excellent, good, medium, poor, and extremely poor groundwater quality, respectively.

Table 3. Quality classification of phreatic water and confined water.

	Categories	Ranges	NO. of Sample
WQI	Excellent	<50	15
	Good	50–100	24
	Medium	100–200	18
	Poor	200–300	3
	Extremely poor	>300	6

shows the distribution of groundwater sampling points in the Shizuishan area among these five categories. There are 15 groundwater samples of excellent quality, which are mainly distributed along the Helan Mountains. This may be related to the good quality of lateral inflow in the groundwater system.

The aquifer along the Helan Mountains consists mainly of gravels, with much higher hydraulic gradients and faster groundwater flow rates. Considering the groundwater recharge sources, these samples may be associated with the good quality of lateral inflow in the groundwater system [39]. In total, 24 and 18 samples were classified as good and medium, respectively. This implies that among the sampling points, 60% of them contain groundwater that is suitable for drinking. In addition, there were three samples of poor quality. Six samples (Q1, Q5, Q8, Q11, Q19, and Q41) were of extremely poor quality. These samples were not suitable for direct drinking. For example, the maximum WQI value in Q8 was 965. The concentrations of TDS, NH4-N, and As in Q8 were 18448, 3.52, and 0.0253 mg/L, respectively, which far exceeded the WHO’s [51] acceptable limits for drinking. The western and southern parts of the Shizuishan region had good groundwater quality, including Chengguan, Touzha, the northern part of Hongguozi, and the eastern part of Chonggang (Figure 9).

4.5. Assessment of the Hazard to Human Health

The non-carcinogenic health risk levels of groundwater oral intake were calculated. The health risks of NH₄-N, NO₃⁻N, F⁻, and As intake for children and adults are shown in Figure 10. The hazard quotients of NH₄-N were in the range of 0–0.22 for children and 0–0.11 for adults. The non-carcinogenic risks posed by NO₃-N and F^- were 0–1.81 and 0.05–1.81 for children, respectively. However, the corresponding risks in adults were acceptable in the range of 0–0.93 for NO₃-N and 0.02–0.62 for F⁻. The non-carcinogenic risks associated with As in groundwater were predominant in the study region. The HQ values of As were between 0 and 20.38 for children, and between 0 and 10.48 for adults. Thus, for children, the cumulative hazard index (HI) through drinking water ranged from 0.09 to 20.65 (mean of 1.53), whereas the maximum was recorded at Q10 (20.65). The HI for adults ranged from 0.05 to 10.62, with a mean of 0.79, which is within the permissible risk level of 1.0. The level of non-carcinogenic hazards posed to children were 1.94 times higher than those posed to adults. In this regard, pollutants in groundwater posed higher non-carcinogenic risks to children. Similar results have been observed by other scholars. Adimalla et al. [43] identified the great potential risks in the agricultural area of Nanganur in southern India due to high concentrations of nitrate in groundwater. Qaiser et al. [47] also found that the non-carcinogenic index was higher for children than for adults in Pakistan.

As shown in

Table 4. Non-carcinogenic and carcinogenic risk results of Shizuishan area through drinking water intake.

		HI			RAs
		NO. of Samples	Percentage of HQ > 1	Contribution rate of HI	NO. of Samples	Percentage of RAs > 1 × 10−4
Children	NH4-N	0	0%	1.61%	/	/
	NO3-N	1	1.52%	11.42%	/	/
	F−	13	19.70%	29.91%	/	/
	As	12	18.18%	57.07%	6	9.09%
	All	29	43.94%	/	/	/
Adults	NH4-N	0	0%	1.61%	/	/
	NO3-N	0	0%	11.42%	/	/
	F−	0	0%	29.91%	/	/
	As	9	13.64%	57.07%	12	18.18%
	All	12	18.18%	/	/	/

, the contribution of HI was 57.07% for As, 29.91% for F⁻, 11.42% for NO₃-N, and 1.61% for NH₄-N for both children and adults. The ranking of groundwater parameters according to their non-carcinogenic risks to human health was As > F^- > NO₃-N > NH₄-N. This also implies that concentrations of As and F have a great influence on the non-carcinogenic risks of groundwater. The spatial distribution of HI had similar characteristics to the spatial distribution of As and F⁻. Groundwater is the only source of drinking water in the study region, and most of the residents, especially children, are exposed to adverse health risks due to the consumption of contaminated groundwater. In particular, the metabolic capacity of children is far lower than that of adults, and the harm from drinking contaminated water is much greater than that posed to adults. The long-term drinking of high-arsenic water can lead to chronic arsenic poisoning, including skin lesions, diabetes, cardiovascular diseases, and cancer [85,86]. Drinking high-fluoride groundwater will cause dental fluorosis and skeletal fluorosis [87]. Although the hydrochemical groundwater data showed that nitrogen contamination was more severe in the study region, the As concentration and the associated risks to residents should be given more attention by decision makers.

For children (Figure 11a), the carcinogenic risk index of As by oral intake was 0–7.37 × 10⁻⁴, with a mean value of 3.02 × 10⁻⁵. For adults (Figure 11b), the carcinogenic risk index of As ranged between 0–1.89 × 10⁻⁴ with a mean level of 7.76 × 10⁻⁵. There were 6 and 12 samples in the Middle East region that exceeded the acceptable level for children and adults, respectively. It may have adverse effects on human health.

4.6. Recommendation for Groundwater Management

Based on the above discussion, serious potential health risks exist in drinking groundwater from the study region. It is consistent with the previous findings [44,64] that the nitrogen pollution of groundwater contamination has become a serious challenge in the Yinchuan Plain [46,66]. The Jing Hui canal irrigation area is another old agricultural region with more than 2000 years of irrigation history. Similarly, nitrate pollution and the related health risks caused by intensive agricultural activities are very serious in the irrigation area [88]. In general, NO₃-N is a common component of nitrogen. When groundwater is in a reducing environment, nitrate will be reduced to nitrite nitrogen and then to ammonia nitrogen under the action of anaerobic bacteria [89]. In other words, NH₄-N in groundwater should also be paid more attention. However, there is an inconsistency between the dominant pollutants and the health-risk-induced pollutants in groundwater. In the study area, the exceedance rate of arsenic in groundwater was 15%, but it accounted for 57.7% of the total non-carcinogenic risk and was the main carcinogenic contaminant causing health risks. Therefore, arsenic contamination in groundwater should be of concern. These high-arsenic groundwaters in China are mainly located in the fluvial/alluvial-lacustrine plains and basins located in arid/semi-arid regions and alluvial plains/basins and river deltas in humid/semi-humid regions [87]. Most areas of the world are at risk from excessive arsenic levels, for example, 59 endemic arsenic villages with approximately 215,600 people at high risk have been identified among the inhabitants of the Datong Basin [24]; approximately 1.56 to 19.8 million people in Bangladesh are thought to be at risk from drinking arsenic-contaminated water [90]. Arsenic contamination is more severe in the Ganges Basin, where several people have developed skin and nerve cancer and most children have been diagnosed with arsenic poisoning [91].

In arid and semi-arid areas, groundwater is an important source of drinking water for domestic use. In the study region, almost every household relies on private wells for drinking water because of the low cost. This means that there is an urgent need to improve public awareness and to understand the relationship between oral ingestion and adverse health effects. The timely disclosure of this information is an effective management approach. Maintaining the safety of drinking water should be the highest priority for the decision makers. Due to the potential adverse health effects, groundwater abstracted from private wells is not recommended for drinking purposes in the study region. Compared with phreatic water, confined water is of a relatively good quality for the multi-layered aquifer system in Yinchuan Plain [92]. However, due to intensive pumping activities, the aquifer leakage between the phreatic aquifer and the confined aquifer caused the downward migration of pollutants. As reported by Chen, high NH₄-N concentrations were detected in confined water for pumping wells, while the corresponding contaminant concentration was below the detection limit for water in monitoring wells [40]. Therefore, a well-developed groundwater monitoring network is of great importance to identify the variation of chemical constituents in groundwater and to elucidate the migration of contaminants. It will be helpful to optimize groundwater abstraction and provide more insight for further management.

5. Conclusions

This study assessed the groundwater quality and hydrochemistry and the associated risks to human health in the Shizuishan area. The results showed that the groundwater was slightly alkaline. TDS values ranged from 232 to 18448 mg/L, with half of the samples exceeding the permissible limit for drinking water. The ranking of groundwater cations according to mean content was Na⁺ > Ca²⁺ > Mg²⁺ > K⁺, while that of anions was SO₄²⁻ > Cl⁻ > HCO₃⁻. Approximately 20% and 5% of the samples had NO₃-N concentrations above the WHO and Chinese drinking water standards, respectively. Nearly 25% of samples had high NH₄-N concentrations above the drinking water threshold and showed a wide spatial distribution. Among the groundwater samples, the fluoride concentrations of 2 samples and 16 samples were higher than the WHO and Chinese standards for drinking water, respectively. While the measured water quality parameters of most groundwater samples were within the permissible limits for drinking water, 10 samples from the central region had levels of As concentrations above the permissible limit. The groundwater hydrochemical facies were mainly SO₄·Cl-Ca·Mg and SO₄·Cl-Na. Based on the principal component analysis, four principal components in groundwater were the salinity factor, carbonate factor, alkaline factor, and pollution factor in controlling groundwater chemistry in the study region.

The value of the WQI was in the range of 28–968. The WQI of about 60% of the groundwater samples indicated the suitability of their use as drinking water. The water samples of excellent quality are distributed along the Helan Mountains. This may be related to the good quality of lateral recharge in the groundwater system. There were three samples of poor quality. Six samples (Q1, Q5, Q8, Q11, Q19, and Q41) were of extremely poor quality. These samples were not suitable for direct drinking. The HI values in the study were in the range 0.05–10.62 for adults and 0.09–20.65 for children, respectively. Children showed a high risk of exposure to groundwater contaminants upon the long-term consumption of contaminated drinking water. The impact of the pollutants in the groundwater for residents decreased in the order As > F⁻ > NO₃-N > NH₄-N. Meanwhile, the carcinogenic risk value of the oral intake of As was 0–7.37 × 10⁻⁴ for children and 0–1.89 × 10⁻⁴ for adults. There were 6 and 12 samples located in the middle east area with levels beyond the permissible acceptable levels. This study can further support groundwater management in the study area. More effective approaches should be considered in the future.